{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,6]],"date-time":"2026-11-06T18:01:04Z","timestamp":1793988064959,"version":"4.8.0"},"reference-count":46,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2022,11,10]],"date-time":"2022-11-10T00:00:00Z","timestamp":1668038400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,11,10]],"date-time":"2022-11-10T00:00:00Z","timestamp":1668038400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Brain tumor is a life-threatening disease and causes about 0.25 million deaths worldwide in 2020. Magnetic Resonance Imaging (MRI) is frequently used for diagnosing brain tumors. In medically underdeveloped regions, physicians who can accurately diagnose and assess the severity of brain tumors from MRI are highly lacking. Deep learning methods have been developed to assist physicians in detecting brain tumors from MRI and determining their subtypes. In existing methods, neural architectures are manually designed by human experts, which is time-consuming and labor-intensive. To address this problem, we propose to automatically search for high-performance neural architectures for classifying brain tumors from MRIs, by leveraging a Learning-by-Self-Explanation (LeaSE) architecture search method. LeaSE consists of an explainer model and an audience model. The explainer aims at searching for a highly performant architecture by encouraging the architecture to generate high-fidelity explanations of prediction outcomes, where explanations\u2019 fidelity is evaluated by the audience model. LeaSE is formulated as a four-level optimization problem involving a sequence of four learning stages which are conducted end-to-end. We apply LeaSE for MRI-based brain tumor classification, including four classes: glioma, meningioma, pituitary tumor, and healthy, on a dataset containing 3264 MRI images. Results show that our method can search for neural architectures that achieve better classification accuracy than manually designed deep neural networks while having fewer model parameters. For example, our method achieves a test accuracy of 90.6% and an AUC of 95.6% with 3.75M parameters while the accuracy and AUC of a human-designed network\u2014ResNet101\u2014is 84.5% and 90.1% respectively with 42.56M parameters. In addition, our method outperforms state-of-the-art neural architecture search methods.<\/jats:p>","DOI":"10.1038\/s41598-022-22172-6","type":"journal-article","created":{"date-parts":[[2022,11,10]],"date-time":"2022-11-10T15:04:44Z","timestamp":1668092684000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":64,"title":["Brain tumor classification based on neural architecture search"],"prefix":"10.1038","volume":"12","author":[{"given":"Shubham","family":"Chitnis","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ramtin","family":"Hosseini","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pengtao","family":"Xie","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2022,11,10]]},"reference":[{"key":"22172_CR1","unstructured":"https:\/\/www.cancer.net\/cancer-types\/brain-tumor\/statistics. ."},{"key":"22172_CR2","doi-asserted-by":"publisher","first-page":"1953","DOI":"10.1002\/cncr.23741","volume":"113","author":"ML Bondy","year":"2008","unstructured":"Bondy, M. L. et al. Brain tumor epidemiology: consensus from the brain tumor epidemiology consortium. Cancer 113, 1953\u20131968 (2008).","journal-title":"Cancer"},{"key":"22172_CR3","doi-asserted-by":"publisher","first-page":"97","DOI":"10.1007\/s00401-007-0243-4","volume":"114","author":"DN Louis","year":"2007","unstructured":"Louis, D. N. et al. The 2007 who classification of tumours of the central nervous system. Acta Neuropathol. 114, 97\u2013109. https:\/\/doi.org\/10.1007\/s00401-007-0243-4 (2007).","journal-title":"Acta Neuropathol."},{"key":"22172_CR4","doi-asserted-by":"publisher","first-page":"8","DOI":"10.14791\/btrt.2015.3.1.8","volume":"3","author":"MC Mabray","year":"2015","unstructured":"Mabray, M. C., Barajas, R. F. & Cha, S. Modern brain tumor imaging. Brain Tumor Res. Treat. 3, 8\u201323 (2015).","journal-title":"Brain Tumor Res. Treat."},{"key":"22172_CR5","doi-asserted-by":"crossref","unstructured":"Afshar, P., Mohammadi, A. & Plataniotis, K.\u00a0N. Brain tumor type classification via capsule networks. In The 25th IEEE international conference on image processing (ICIP), 3129\u20133133 (organizationIEEE, 2018).","DOI":"10.1109\/ICIP.2018.8451379"},{"key":"22172_CR6","doi-asserted-by":"publisher","first-page":"2021","DOI":"10.3389\/frai.2021.635766","volume":"4","author":"R Hao","year":"2021","unstructured":"Hao, R., Namdar, K., Liu, L. & Khalvati, F. A transfer learning-based active learning framework for brain tumor classification. Front. Artif. Intell. 4, 2021 (2021).","journal-title":"Front. Artif. Intell."},{"key":"22172_CR7","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2019.101678","volume":"57","author":"N Ghassemi","year":"2020","unstructured":"Ghassemi, N., Shoeibi, A. & Rouhani, M. Deep neural network with generative adversarial networks pre-training for brain tumor classification based on mr images. Biomed. Signal Process. Control 57, 101678 (2020).","journal-title":"Biomed. Signal Process. Control"},{"key":"22172_CR8","doi-asserted-by":"crossref","unstructured":"D\u00edaz-Pernas, F.\u00a0J., Mart\u00ednez-Zarzuela, M., Ant\u00f3n-Rodr\u00edguez, M. & Gonz\u00e1lez-Ortega, D. A deep learning approach for brain tumor classification and segmentation using a multiscale convolutional neural network. In Healthcare, vol.\u00a09, 153 (organizationMultidisciplinary Digital Publishing Institute, 2021).","DOI":"10.3390\/healthcare9020153"},{"key":"22172_CR9","unstructured":"Zoph, B. & Le, Q.\u00a0V. Neural architecture search with reinforcement learning. In International Conference on Learning Representations (2017)."},{"key":"22172_CR10","unstructured":"Pham, H., Guan, M.\u00a0Y., Zoph, B., Le, Q.\u00a0V. & Dean, J. Efficient neural architecture search via parameter sharing. In International Conference on Machine Learning (2018)."},{"key":"22172_CR11","doi-asserted-by":"crossref","unstructured":"Zoph, B., Vasudevan, V., Shlens, J. & Le, Q.\u00a0V. Learning transferable architectures for scalable image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2018).","DOI":"10.1109\/CVPR.2018.00907"},{"key":"22172_CR12","unstructured":"Liu, H., Simonyan, K. & Yang, Y. DARTS: differentiable architecture search. In International Conference on Learning Representations (2019)."},{"key":"22172_CR13","unstructured":"Such, F.\u00a0P., Rawal, A., Lehman, J., Stanley, K. & Clune, J. Generative teaching networks: Accelerating neural architecture search by learning to generate synthetic training data. In International Conference on Machine Learning, 9206\u20139216 (organizationPMLR, 2020)."},{"key":"22172_CR14","unstructured":"Liu, H., Simonyan, K., Vinyals, O., Fernando, C. & Kavukcuoglu, K. Hierarchical representations for efficient architecture search. In International Conference on Learning Representations (2018)."},{"key":"22172_CR15","doi-asserted-by":"publisher","first-page":"4780","DOI":"10.1609\/aaai.v33i01.33014780","volume":"33","author":"E Real","year":"2019","unstructured":"Real, E., Aggarwal, A., Huang, Y. & Le, Q. V. Regularized evolution for image classifier architecture search. In Proceedings of the AAAI conference on artificial intelligence 33, 4780\u20134789 (2019).","journal-title":"In Proceedings of the AAAI conference on artificial intelligence"},{"key":"22172_CR16","unstructured":"Cai, H., Zhu, L. & Han, S. Proxylessnas: Direct neural architecture search on target task and hardware. In International Conference on Learning Representations (2019)."},{"key":"22172_CR17","unstructured":"Chen, X. & Hsieh, C.-J. Stabilizing differentiable architecture search via perturbation-based regularization. In International Conference on Machine Learning, 1554\u20131565 (organizationPMLR, 2020)."},{"key":"22172_CR18","unstructured":"Chu, X. et\u00a0al. Darts-: Robustly stepping out of performance collapse without indicators. In International Conference on Learning Representations (2020)."},{"key":"22172_CR19","unstructured":"Zela, A. et\u00a0al. Understanding and robustifying differentiable architecture search. In International Conference on Learning Representations (2020)."},{"key":"22172_CR20","unstructured":"Hosseini, R. & Xie, P. Learning by self-explanation, with application to neural architecture search. arXiv preprint arXiv:2012.12899 (2020)."},{"key":"22172_CR21","unstructured":"Goodfellow, I.\u00a0J., Shlens, J. & Szegedy, C. Explaining and harnessing adversarial examples. In International Conference on Learning Representations (2015)."},{"key":"22172_CR22","unstructured":"Etmann, C., Lunz, S., Maass, P. & Schoenlieb, C. On the connection between adversarial robustness and saliency map interpretability. In International Conference on Machine Learning, 1823\u20131832 (organizationPMLR, 2019)."},{"key":"22172_CR23","doi-asserted-by":"publisher","first-page":"317","DOI":"10.1016\/j.procs.2016.09.407","volume":"102","author":"A I\u015f\u0131n","year":"2016","unstructured":"I\u015f\u0131n, A., Direko\u011flu, C. & \u015eah, M. Review of mri-based brain tumor image segmentation using deep learning methods. Procedia Comput. Sci. 102, 317\u2013324 (2016).","journal-title":"Procedia Comput. Sci."},{"key":"22172_CR24","doi-asserted-by":"publisher","first-page":"1993","DOI":"10.1109\/TMI.2014.2377694","volume":"34","author":"BH Menze","year":"2014","unstructured":"Menze, B. H. et al. The multimodal brain tumor image segmentation benchmark (brats). IEEE Trans. Med. Imaging 34, 1993\u20132024 (2014).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"22172_CR25","doi-asserted-by":"publisher","first-page":"1240","DOI":"10.1109\/TMI.2016.2538465","volume":"35","author":"S Pereira","year":"2016","unstructured":"Pereira, S., Pinto, A., Alves, V. & Silva, C. A. Brain tumor segmentation using convolutional neural networks in mri images. IEEE Trans. Med. Imaging 35, 1240\u20131251 (2016).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"22172_CR26","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1016\/j.media.2016.05.004","volume":"35","author":"M Havaei","year":"2017","unstructured":"Havaei, M. et al. Brain tumor segmentation with deep neural networks. Med. Image Anal. 35, 18\u201331 (2017).","journal-title":"Med. Image Anal."},{"key":"22172_CR27","doi-asserted-by":"publisher","first-page":"90","DOI":"10.1016\/j.patcog.2018.11.009","volume":"88","author":"S Chen","year":"2019","unstructured":"Chen, S., Ding, C. & Liu, M. Dual-force convolutional neural networks for accurate brain tumor segmentation. Pattern Recognit. 88, 90\u2013100 (2019).","journal-title":"Pattern Recognit."},{"key":"22172_CR28","doi-asserted-by":"publisher","first-page":"174","DOI":"10.1016\/j.jocs.2018.12.003","volume":"30","author":"M Sajjad","year":"2019","unstructured":"Sajjad, M. et al. Multi-grade brain tumor classification using deep cnn with extensive data augmentation. J. Comput. Sci. 30, 174\u2013182 (2019).","journal-title":"J. Comput. Sci."},{"key":"22172_CR29","doi-asserted-by":"crossref","unstructured":"Kaldera, H., Gunasekara, S.\u00a0R. & Dissanayake, M.\u00a0B. Brain tumor classification and segmentation using faster r-cnn. In Advances in Science and Engineering Technology International Conferences, 1\u20136 (organizationIEEE, 2019).","DOI":"10.1109\/ICASET.2019.8714263"},{"key":"22172_CR30","doi-asserted-by":"crossref","unstructured":"Ghosal, P. et\u00a0al. Brain tumor classification using resnet-101 based squeeze and excitation deep neural network. In 2019 Second International Conference on Advanced Computational and Communication Paradigms (ICACCP), 1\u20136 (organizationIEEE, 2019).","DOI":"10.1109\/ICACCP.2019.8882973"},{"key":"22172_CR31","doi-asserted-by":"publisher","first-page":"903","DOI":"10.1007\/s10278-020-00347-9","volume":"33","author":"H Mzoughi","year":"2020","unstructured":"Mzoughi, H. et al. Deep multi-scale 3d convolutional neural network (cnn) for mri gliomas brain tumor classification. J. Digit. Imaging 33, 903\u2013915 (2020).","journal-title":"J. Digit. Imaging"},{"key":"22172_CR32","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41598-020-74419-9","volume":"10","author":"L Pei","year":"2020","unstructured":"Pei, L., Vidyaratne, L., Rahman, M. M. & Iftekharuddin, K. M. Context aware deep learning for brain tumor segmentation, subtype classification, and survival prediction using radiology images. Sci. Rep. 10, 1\u201311 (2020).","journal-title":"Sci. Rep."},{"key":"22172_CR33","doi-asserted-by":"crossref","unstructured":"Shaik, N.\u00a0S. & Cherukuri, T.\u00a0K. Multi-level attention network: application to brain tumor classification. Signal, Image and Video Processing 1\u20138 (2021).","DOI":"10.1007\/s11760-021-02022-0"},{"key":"22172_CR34","doi-asserted-by":"crossref","unstructured":"Lu, D., Polomac, N., Gacheva, I., Hattingen, E. & Triesch, J. Human-expert-level brain tumor detection using deep learning with data distillation and augmentation. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 3975\u20133979 (organizationIEEE, 2021).","DOI":"10.1109\/ICASSP39728.2021.9415067"},{"key":"22172_CR35","doi-asserted-by":"publisher","first-page":"1655","DOI":"10.1002\/ima.22543","volume":"31","author":"S Deepak","year":"2021","unstructured":"Deepak, S. & Ameer, P. Brain tumour classification using siamese neural network and neighbourhood analysis in embedded feature space. Int. J. Imaging Syst. Technol. 31, 1655\u20131669 (2021).","journal-title":"Int. J. Imaging Syst. Technol."},{"key":"22172_CR36","unstructured":"Wang, F. Neural architecture search for gliomas segmentation on multimodal magnetic resonance imaging. arXiv preprint arXiv:2005.06338 (2020)."},{"key":"22172_CR37","doi-asserted-by":"crossref","unstructured":"Milesi, A., Futrega, M., Marcinkiewicz, M. & Ribalta, P. Brain tumor segmentation using neural network topology search. In International MICCAI Brainlesion Workshop, 366\u2013376 (organizationSpringer, 2022).","DOI":"10.1007\/978-3-031-08999-2_31"},{"key":"22172_CR38","unstructured":"Xie, S., Zheng, H., Liu, C. & Lin, L. SNAS: stochastic neural architecture search. In International Conference on Learning Representations (2019)."},{"key":"22172_CR39","doi-asserted-by":"crossref","unstructured":"Chen, X., Xie, L., Wu, J. & Tian, Q. Progressive differentiable architecture search: Bridging the depth gap between search and evaluation. In International Conference on Computer Vision (2019).","DOI":"10.1109\/ICCV.2019.00138"},{"key":"22172_CR40","unstructured":"Xu, Y. et\u00a0al. PC-DARTS: partial channel connections for memory-efficient architecture search. In International Conference on Learning Representations (2020)."},{"key":"22172_CR41","doi-asserted-by":"publisher","unstructured":"Bhuvaji, S., Kadam, A., Bhumkar, P., Dedge, S. & Kanchan, S. Brain tumor classification (mri). Kaggle https:\/\/doi.org\/10.34740\/KAGGLE\/DSV\/1183165 (2020).","DOI":"10.34740\/KAGGLE\/DSV\/1183165"},{"key":"22172_CR42","doi-asserted-by":"publisher","first-page":"S2","DOI":"10.1188\/16.CJON.S1.2-8","volume":"20","author":"ME Davis","year":"2016","unstructured":"Davis, M. E. Glioblastoma: Overview of disease and treatment. Clin. J. Oncol. Nurs. 20, S2\u2013S8. https:\/\/doi.org\/10.1188\/16.CJON.S1.2-8 (2016).","journal-title":"Clin. J. Oncol. Nurs."},{"key":"22172_CR43","doi-asserted-by":"crossref","unstructured":"Cubuk, E.\u00a0D., Zoph, B., Mane, D., Vasudevan, V. & Le, Q.\u00a0V. Autoaugment: Learning augmentation policies from data. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2019).","DOI":"10.1109\/CVPR.2019.00020"},{"key":"22172_CR44","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 770\u2013778 (2016).","DOI":"10.1109\/CVPR.2016.90"},{"key":"22172_CR45","unstructured":"Simonyan, K. & Zisserman, A. Very deep convolutional networks for large-scale image recognition. In International Conference on Learning Representation (2014)."},{"key":"22172_CR46","doi-asserted-by":"crossref","unstructured":"Huang, G., Liu, Z., van\u00a0der Maaten, L. & Weinberger, K.\u00a0Q. Densely connected convolutional networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2017).","DOI":"10.1109\/CVPR.2017.243"}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-022-22172-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-022-22172-6","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-022-22172-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,11,10]],"date-time":"2022-11-10T15:12:47Z","timestamp":1668093167000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-022-22172-6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,11,10]]},"references-count":46,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2022,12]]}},"alternative-id":["22172"],"URL":"https:\/\/doi.org\/10.1038\/s41598-022-22172-6","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,11,10]]},"assertion":[{"value":"15 January 2022","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 October 2022","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"10 November 2022","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"19206"}}